With to Dan Pharis, I will disagree in regards to the preload on main springs. The MAIN criteria for gaining mechanical advantage on tumbler/cock is positioning the "hook" of the mainspring as close as is physically possible to the axel of the lock plate/tumbler bushing! as this "hook" rides up the tumbler shoe towards the axel the cock pressure should lesson because of the Physics known as MA. H.T.
Thanks for the reply, Hugh. I'm sorry if my earlier reply sounded like I was disagreeing with you. Not my intention. I agree completely that the main reason the felt load decreases is that the tip of the spring goes toward the axle as the hook rides up the cam of the shoe. Yes, that's how it gains mechanical advantage.
I know you understand this, Hugh. But for the benefit of anyone reading this who doesn't understand it, it's just like when we used to move up and sit closer to the fulcrum on the seesaw so that we could ride with someone lighter than ourselves. In terms of physics, we were shortening the lever on our side of the see saw. The spring does the same as it moves closer and closer to the axle, as the cock is drawn back. It's just like us moving closer and closer to the fulcrum on the seesaw.
Back to Dan. Dan's post was about making springs, and more specifically, about the benefits of the preload (downward) bend. If I remember correctly, his point was that preload not only adds a little strength to the spring (and, as some suggest, it looks nice for the spring to be straight when at full cock). It also helps the spring work at its best for mechanical advantage. Springs will drive a lock whether or not they are made perfectly or mounted perfectly. But in concept, when a spring is straight, it will at that point be its longest, from bend to tip. So if the spring is made in such a way that the lower limb is straight when compressed, then when it is compressed (in a spring vise) it's near to its longest and it's easy to position it so as to have the tip as close as possible to the tumbler axle at full cock (as we agree it should be for optimum mechanical advantage). (I say "near to it's longest", because a mainspring vise only compresses a spring opposite the top leaf. So the profile of a spring compressed in a vise is still slightly different from a spring in a lock pulled to full cock.) A spring that is straight at its compressed, full-cock position will naturally move away from the tumbler axle as it relaxes.
For anyone unfamiliar with this, I should point out that the Siler springs do have a preload.
Also, this is not suggesting that springs that have an upward bend at full cock are inferior. They work fine, and some antique locks and modern locks were and are made that way.